Item

Abstract

Computing global illumination (GI) in virtual scenes becomes increasingly attractive even for real-time applications nowadays. GI delivers important cues in the perception of 3D virtual scenes, which is important for material and architectural design. Therefore, for photo-realistic rendering in the design and even the game industry, GI has become indispensable. While the computer simulation of realistic global lighting is well-studied and often considered as solved, computing it efficiently is not. Saving computation costs is therefore the main motivation of current research in GI. Efficient algorithms have to take various aspects into account, such as the algorithmic complexity and convergence, its mapping to parallel processing hardware, and the knowledge of certain lighting properties including the capabilities of the human visual system. In this dissertation we exploit both low-level and high-level coherence in the practical design of GI algorithms for a variety of target applications ranging from high-quality production rendering to dynamic real-time rendering. We also focus on automatic rendering-accuracy control to approximate GI in such a way that the error is perceptually unified in the result images, thereby taking not only into account the limitations of the human visual system but also later video compression with an MPEG encoder. In addition, this dissertation provides many ideas and supplementary material, which complements published work and could be of practical relevance.